Presentation with ‘live’ musical examples (John Bryan, Alison Crum, Roy Marks) given at Pan-Pacific Gamba Gathering, University of Hawaii at Manoa, Honolulu, Hawaii 1 August 2007 John Bryan ‘Verie sweete and artificiall’: Lorenzo Costa and the earliest viols The importance of the courts of Northern Italy as the birthplace of the viol consort in the closing years of the fifteenth century has been well established since the 1980s,1 but until recently any performances of the repertory associated with the fledgling viol consort, such as the frottole, Franco-Flemish chansons and ‘songs without words’2 preserved in publications such as Petrucci’s Odhecaton of 1501 and his series of frottola books from 1504, and contemporary manuscript collections, have been played on instruments (or rather copies of them) dating from considerably later than the 1490s and early 1500s. Most frequently, reproductions of Venetian viols from the second half of the 16th century have been utilised, but the appearance and construction of these instruments is markedly different from those shown in iconographical sources from the period actually contemporary with this repertory.3 One such early model, identified by Ian Woodfield in his 1984, The Early History of the Viol,4 is Lorenzo Costa’s Madonna and Child enthroned with SS. Augustine, Posidonius, John and Francis, the altarpiece commissioned by Francesco di Giacomo Ghedini for his family Cappella di S. Agostino in the church of San Giovanni in Monte, Bologna (illus. 1). This painting, securely dated 1497 on the lintel below the Madonna’s feet, is by an artist trained in Ferrara and directly associated later in his career with the court of Isabella d’Este (1474–1539), marchioness of Mantua from 1490, whose patronage of the early viol was a considerable force in its early development. The purpose of this presentation is to highlight the significance of the viols depicted in Costa’s altarpiece to our understanding of the early viol consort by assessing the painting’s relevance as a source of information, to investigate its relationship to other documentary evidence concerning the earliest development of the viol consort, and to evaluate the results of performing on a set of instruments constructed on the basis of this painting. Any hypothetical construction of instruments from an iconographical source must tread very cautiously, addressing questions of the relevance and reliability of the artwork concerned. How ‘good’ a source is Costa’s Ghedini altarpiece in providing information about a putative early viol consort? What were Costa’s connections with professional musicians and their amateur patrons? How closely would he have been aware of the development of ‘new’ instruments such as the viols played at the Northern Italian courts like those of the d’Este in Ferrara and the Gonzaga in Mantua? A summary of Lorenzo Costa’s career gives clear evidence that he did indeed operate in just the same artistic circles as those patrons who were instigating the development of the viol consort.5 Born in Bologna around 1460,6 Costa trained in Ferrara with Francesco Cossa and Ercole de Roberti (who worked both at the d’Este court and also for the Gonzaga family at Mantua following Isabella d’Este’s marriage to Francesco Gonzaga in 1490). During the 1480s and 1490s Costa received a series of commissions from the ruling Bentivoglio family of Bologna. The younger members of the Bentivoglio family were assiduous amateurs of music,7 and it is possible that Costa’s well-known Concert in the National Gallery, London (illus. 2),8 with its acutely observed depiction of a trio of singers with lute, may show some of these in performance. The connections between the Bentivoglio family and those of the d’Este of Ferrara and the Gonzagas of Mantua had many strands: from 1474 Giovanni Bentivoglio was an ally and confidant of Isabella’s father Duke Ercole d’Este; in 1487 Lucrezia d’Este married Giovanni Bentivoglio’s eldest son Annibale; while in 1494 Giovanni Gonzaga married Laura, the sixth daughter of Giovanni Bentivoglio. In 1504 when Isabella d’Este was seeking a painting for her studiolo by a leading Bolognese artist, it was to Antongaleazzo Bentivoglio, a discriminating patron of the arts and scholarship, that she turned for advice, and who helped to provide Lorenzo Costa to fulfil the commission that resulted in the Coronation of a Lady9 (illus. 3). On 16 November 1506 Costa joined the Gonzagas at Mantua as court painter in place of Andrea Mantegna who had died in September. Costa became a citizen of Mantua in 1509, and by 1512 had completed his second painting for Isabella’s studiolo, the Kingdom of Comus.10 As we shall, music played almost as great a part in Isabella’s cultural life as did her dedication to the decorative schemes for her private chambers in the ducal palace at Mantua. The two elements come together in the marquetry intarsias that she installed in her camerini, since one panel is incribed with Ockeghem’s three-part canon: Prenez sur moy votre example, representing the ‘learned’ Flemish style so admired by Italian patrons of music. Example 1: Ockeghem: Prenez sur moy votre example. Although The Concert is the only surviving painting by Costa in which a musical performance is obviously the subject,11 music or musical instruments are included in a significant number of his allegorical works, portraits and altarpieces. Angelic musicians were a common decorative feature in Renaissance religious painting: Costa includes them in a number of his works. Bowed string instruments painted by Costa include the straight-sided four- stringed, flat-bridged kit on the parapet in front of the singers in The Concert with its precisely delineated bow, its hair tied at one end and looped at the other. Depicted in rather less detail is a series of lire da braccio beginning with the 1490 Triumph of Death fresco in the Bentivoglio chapel in San Giacomo, Bologna, showing the back of a figure-of-eight shaped lira.12 Both Costa’s allegories for Isabella’s studiolo in Mantua show lire da braccio. In The Coronation of a Lady Apollo/Orpheus symbolically represents the combined rhetorical power of poetry and music: his instrument has one of the shapes most commonly shown in Italian paintings from the first third of the sixteenth century,13 characterised by only one set of corners on the lower bouts, and concave ribs. The association of this particular body shape with the Mantuan court (several commentators have seen this painting as a depiction of Isabella herself, surrounded by musicians and poets)14 is a significant element in the search for the early viol, since it is identical to that of the viols shown in Costa’s Ghedini altarpiece. The inclusion of angelic musicians in images of the Madonna enthroned was already well established by the late 15th century, with the lira da braccio and lute the most regularly depicted instruments. A particular feature of altarpieces by Ferrarese artists towards the end of the century was the elevation of Mary’s throne, creating a space below it that gains prominence and therefore significance from its proximity to the viewer, on whose eye- level it is situated.15 This space was often filled by Costa and his contemporaries with the beautifully detailed distant landscapes for which they were particularly renowned. In the 1497 Ghedini altarpiece just such a graceful landscape is placed behind the conventional image of two angelic musicians. However the nature of the instruments they play is unique in paintings in this genre. Costa has here placed a pair of viols in a position of particular importance and high status in which they become a secondary focal point of attention very different from the marginal musicians in several of his other paintings. Costa’s inclusion of viols in such a prestigious position certainly suggests that he was fully aware that these novel instruments were beginning to join the lute and lira as participants in the sophisticated music of the courts of the patrons for whom he worked, and was consciously drawing the viewer’s attention towards them. The area of the painting containing the two angelic-looking youngsters with their viols measures approximately 56 x 43 cm (illus. 4). Especially in such a relatively small image, it must be accepted that there are significant dangers in building too detailed a hypothesis from iconographic detail created at a time when the art of natural perspective was still in its early development, and when one might easily question the accuracy of the instruments depicted.16 Measurements are particularly difficult to pin down, as we have here no absolute standard from which to work. However, it is clear that Costa has set out to show two viols of different sizes: the body of the larger viol is approximately 10 per cent longer than that of the smaller, but this masks the true difference in size. The larger viol has a comparatively longer neck in comparison to its body size, accounting for a nut–tail measurement exceeding that of the smaller viol by approximately 17 per cent, and when its lower bridge position is taken into account, it has a string length some 22 per cent longer than its smaller partner.17 This would make it capable of sounding a fourth or fifth lower and it would thus be reasonable to consider these as a tenor and bass (or possibly treble and tenor) pair of viols, confirming their function as instruments designed to play polyphonic strands of different ranges. In contrast, the lira da braccio is usually shown in iconography as a solo instrument, commonly with a flatter bridge and bourdon strings, which is unsurprising in view of its most frequent function, that of providing a largely chordal accompaniment to a solo vocal improvised recitation. The fronts of Costa’s viols are clearly curved, and thus most likely carved. They have large ‘C’ holes, and show the distinctive lira-like outline with two lower corners only. The backs of the viols are not visible, but the narrow ribs have a noticeably concave profile, with fronts and backs overlapping them. All these features link Costa’s viols closely to the most commonly depicted forms of the lira da braccio from this period, and mark them out from the figure-of-eight pattern, flat fronts and flush-fitted straight-profiled ribs associated with the Valencian vihuela d’arco. The instruments shown in the Ghedini altarpiece are clearly not lire da braccio: they are too large to be played on the arm, and have much longer necks than contemporary lire. Thus it seems likely that Costa’s viols may depict a rival model to those that were being introduced into the Northern Italian courts by Spanish musicians associated with the Borgia family.18 Only the bridge of the smaller (‘tenor’) viol is clearly visible in any detail (and the bass end is obscured by the player’s leg). It is made from a darker wood than the front (or possibly stained to match the dark fingerboard and tailpiece). Both bridges are clearly arched, designed to facilitate playing the single lines of a polyphonic composition rather than the chordal approach of the lira. Six strings pass over the tenor’s bridge, though they are not equally spaced, and only five pegs are visible. It could be construed that the lines representing the fifth and sixth ‘strings’ are in fact Costa’s attempt to depict the outside surfaces of a single fat bass string. Or it could just be a lack of interest in such fine detail. The bass viol definitely has six pegs, but its strings are not clearly decipherable as they are obscured by shadow. Both pegboxes are scrolled (curiously the tenor viol’s follows the pattern that became conventional while its larger partner’s curls in the reverse direction). This clearly distinguishes them from the flat inverted leaf- or heart-shaped pegbox with perpendicular pegs that is typical of the lira da braccio. The bows held by the musicians have long and relatively thick sticks, tapering to elegant points, and black hair. While the tenor player does not yet have his instrument in playing position, the bass player checks his left hand position on the frets and bows the strings, holding the bow in a grip that is not dissimilar to that described by Ganassi in Regula Rubertina (Venice, 1542): the hand is just to the right of the frog and his fingers are not in contact with the hair at all. In order to evaluate the usefulness of Costa’s 1497 painting in understanding the early history of the development of the viol consort, it is important to place it in a chronological context. How does it relate to what we know from documentary evidence about the commissioning and playing of viols in consort in the courts of Northern Italian patrons around that time? The court records of Ferrara show the employment of string players (‘pulsatores lirae’), members of the della Viola clan, from as early as 1466.19 However it is important to treat instrument terminology with care, since the terms viola and lira were often used interchangeably, and might refer to plucked or bowed strings, while the latter might be da braccio or da gamba. By 1473 Johannes Martini (c. 1440–1497) had arrived in Ferrara to direct the court’s music for Duke Ercole. Martini was later a significant musical influence on the young Isabella, teaching her to sing at Ferrara and attending her after her move to Mantua following her marriage in 1490. Martini’s many surviving textless pieces such as La Martinella (ex. 2) are ideal candidates for the repertory enjoyed by the early viol consort. Example 2: Martini: La Martinella (from Florence MS Banco Rari 229, c.1490) In 1491 Isabella’s singing lessons with Martini continued,20 but she also began learning stringed instruments with her own musician (‘nostro sonatore’), Giovanni Angelo Testagrossa (1470–1530).21 Isabella played lira da braccio and lute but does not yet appear to be playing viols. On 6 March 1493 Bernardo de’ Prospero, Isabella’s correspondent in Ferrara, wrote to her describing a performance given by Spanish musicians from Rome who were using ‘viols almost as large as myself’.22 Two years later (and just two years before Costa’s Ghedini altarpiece was painted), in March 1495, Isabella herself began her own viol consort by ordering three viole from an unknown maker that she had visited in Brescia.23 At various times in her correspondence these instruments are referred to as viole grande, violoni, or viole ovver lire, indicating both that their size was a significant feature, as was their relationship (perhaps the body shape) to the lira. Testagrossa appears to have carried considerable authority and exerted some influence upon Isabella’s viol playing. Documented as a member of her court from 1495 to 1501 it was he whom she despatched to assess, negotiate the price and collect the three new Brescian viole as noted in a letter of 19 June 1495.24 Isabella’s instructions to Testagrossa regarding the price of the viols show her business acumen: she only wants to pay 15 ducats, but would raise her offer to 18 ducats if he thinks they are worth the extra.25 In 1497, the very year of Costa’s Ghedini altarpiece, Agostino della Viola joined Andrea and Zampaulo della Viola as a chamber musician at Ferrara, perhaps suggesting the establishment of an enlarged string consort.26 Meanwhile in Mantua Isabella received a gift of two viole from Hungary: if they were built as a pair of instruments, they are more likely to have been viols rather than lire.27 The personal involvement of Ercole’s music-loving children in viol playing reached a peak in 1499. On 19 March Lorenzo da Pavia informed Isabella that Alfonso, her younger brother, had requested him to make five viols (viole da archo) ‘in all the possible sizes (modi) in the world’ for him as ‘His Lordship wants to learn [to play them] ... and that with the greatest insistence. It is no use if I say I cannot make them’.28 Does this suggest that Lorenzo was simply too busy, or that he did not know how to make instruments in the required style? Three sizes of viol, tuned a fourth or fifth apart, and five or six instruments could now bring the fuller-scored chansons of Josquin and his contemporaries within the scope of the viol consort’s repertory. Only two months later, on 14 May 1499 Isabella wrote to her half-brother Giulio that she had begun to play viola, had been practising for two days and hoped to ‘play tenor to Don Alfonso’ when she next went to Ferrara.29 Perhaps after further practice she would have been able to tackle Ghiselin’s three-part fantasia possibly named for her brother, La Alfonsina. Example 3: Ghiselin: La Alfonsina (from Odhecaton, 1501) Acquisition of instruments and professional players at both the courts of Ferrara and Mantua went on apace: Jacomo della Viola was first recorded at Ferrara, further augmenting the della Viola consort, while on 22 August 1499 Isabella ordered another viola grande from her maker in Brescia to match her larger viols.30 The order for a large viol not only confirms that the first three were of differing sizes, but that from early in the consort’s development it was the larger sizes that predominated in the consort. It is significant that Isabella went back to her Brescian viol maker in 1499 when she ordered her extra viola grande. She probably realised that the same maker would best be able to construct an additional instrument that would blend well with her existing set. Brescia certainly gained a reputation only a little later in the 16th century as an important centre for the making of viols and violins: Zanetto da Montichiaro (1489/90–1560/1) was noted in the city records as ‘magister a violonis et violis’31 and his instruments were praised by Lanfranco in his Scintille di musica (published in Brescia in 1533). The desire for larger sizes of viols and richer textures reached a climax at the celebrations for the wedding of Alfonso d’Este to Lucrezia Borgia in Ferrara in 1502. The Venetian diarist Marin Sanuto noted six viols being played, one by Alfonso himself during the third act of the festive intermedi following the wedding.32 The viol consort was in regular use by now, and continued to expand in scope: on 5 February 1506 viols, possibly lira-shaped, were reported by Prospero when he wrote from Ferrara to Isabella concerning intermedi in which ‘lire grande’ had been played ‘da octo persone’.33 Isabella continued to sing and play after this, particularly encouraging her Mantuan court composers such as Tromboncino in setting vernacular frottola texts in a more homophonic style than that of Flemish polyphony. Example 4: Bartolomeo Tomboncino: Su su leva However, by around 1505 the focus of her artistic endeavours had turned more toward the visual decoration of her studiolo, overseen now by Costa who had been appointed court painter at Mantua. From this chronological survey it is clear to see that his 1497 Ghedini altarpiece sits fairly and squarely at the beginning of the very period during which the germination of the Italian viol consort was so well documented at the Northern Italian courts. What instrumental properties might have been demanded by the patrons of the early viol consort? Such instruments would have to be capable of playing the repertory for which they appear to have been designed, which, judging by the manuscript collections of North Italian provenance, and Petrucci’s early prints, included chansons in the Burgundian tradition, the newer frottole and the various genres of instrumental ‘songs without words’ which led to the development of the fantasia. This implies a range of sizes appropriate to the tessituras of the polyphonic parts (treble, two more-or-less equally ranged tenors, and bass) with a total notated range of F to f"/g". The different voices of a composition suitable for the incipient viol consort suggest different functions: the cantus is often the most ‘vocal’ in style, tending to move with moderate pace and within a relatively narrow pitch range; the bassus strongly delineates the harmonic framework with more frequent leaps; altus and tenor parts, often working in the same range as each other, tend to be more fluid and decorative, sometimes with syncopations and running passagework. While the tenor requires a good lower register, since it quite frequently supplies the final note at structural cadences while the bassus rests, or leaps over it to the upper fifth, the altus is often a more quixotic voice, and can often be omitted from the texture (as it was in many lute intabulations of frottole) without causing significant damage to the music. Additional si placet parts, most frequently occupy the altus territory as well. This might imply two viols of similar pitch but with slightly different tonal qualities: the one with